Re: Standardization of TeX names for Adobe PostScript fonts.

[bbeeton <BNB@MATH.AMS.COM>]

the original tex on the dec-10 had the 6-character limit.  the sail
computer at stanford ran under waits, but i believe the limit was
the same in tops-10.  it was never a limit in tops-20, which allowed
file names of >30 characters (though i believe early versions were
more efficient if names were unique in the first some small number,
perhaps 6).  i believe there is still one dec-10 in use with tex in
germany; i am not aware of any others.
by the way, the algorithm for deriving 6-character names was not to
take the first 6 characters, but the first 3 and the last 3.  that is
the reason that cm bold math italic is named cmmib* and not cmbmi*,
which would have been the more obvious choice.  the latex fonts do
reduce satisfactorily to 6 characters with this technique.
bart childs once informed me that the limit on length of file names
was even more restrictive on the cray, perhaps 5; he was reduced to
using a code for the size -- a for 10, b for 11, etc., with unlikely
numbers omitted as the size increased.  (i can check on this, if
anyone is interested.)
it's true that the only real reason to use a cray for tex is if you
already happen to be using it for all of your other work, and that
is the basis on which bart said it was actually being used at los
alamos.  under those circumstances, it's highly unlikely that anyone
will install a typesetter or probably even a competent postscript
system on that machine, and that particular restriction can then be
ignored.
the 8-character limit, on the other hand, is ubiquitous on personal
computers, and must therefore be accommodated in some reasonable way.
there is also, of course, the problem of flat file systems on ibm
mainframes and perhaps other machines.  i wouldn't like to see any
"solution" adopted that excluded that part of the tex community.
						-- bb
-------